Firing Angle Document

8/17/2019 Firing Angle Document

1/226

DIGITAL CONTROL O F RECTIFIER FIRING ANGLES FOR THE ZEROGRADIENT SYNCHROTRON (ZGS) RING MAGNET POWER S U P P L Y 5

Martin J. Knott, Lloyd G. Lewis, and Her be rt H. Rabe

Argonne National Laboratory

Argonne, Illinois

Abstract

The hea rt of the new control sy st em for the ZGS

ring magnet power supply is a counter that counts

f r o m 0 to 3600 each voltage cycle of the mai n gene ra-

to r .

synchronized with the g enerat or voltage wave by a

phase - lock feedback loop.

number i n the counter with the desir ed angle forf irin g

each of the 12 rectifier phases.

is generated and applied to the control grids of the

appropriate mercury vapor rectifiers.

arithmetic adder circuits upclate the desired firing

angles for each of the 1 2 phase s in response to com-

mands fr om the ZGS pro gra mm er and in resp ons e t o .

feedback s lgnals fr om beam spill mon itors and f r o m B

pickup co il s on the ring magnet.

arithme tic adders and selecto rs provide individual ad -

justment of each phase in order to reduce low fre -

quency ripple.

l o o - ,

and 150-cycle ripple on the ring magnet field onflattop and on porch es, has provided fas t action to

pe rmi t spill con tr ol when the RF accelerating cavity

is off, and has provided st able opera tion in full rectify

f o r the accelerating pa rt of the ZGS cycle.

This provides an electrica l degree scale that is

Digital gates c ompare the

At eauality, a pulse

Fast digital

Separa te digital

This s ystem has greatly reduced 50 - ,

Introduction

The new slow resonance e xtrac tion s yst em for the

ZGS provides two simultaneous beams that have no R F

s t ructure.

off the R F accel erati ng cavity ea rl y in flattop and then,

after the beam has debunched, by moving the beam t o

the v r = 2 / 3 extraction point with the ring magnet con-

tro l system. The rate of extraction is then controlled

throu gh manipulation of the rin g nlagnet field.

The sys tem accomplishes this by turning

This magnetic spil l mode place s specific demands

on the con trol sys tem for the ring magnet power sup-

ply. One of the se demands is that the s yst em must beable to rapi dly change the voltage applied to the rin g

magnet. The required speed and accuracy ar e det er -mined by the amount of beam s teer ing needed to s t a r t

and maintain the extraction at a select ed rat e.

A second demand is that the cont rol sy st em for

the ring magn et minimizes the g enera tion of low fr e -

quency ripple. This is important becau se the ripple

produces modu la ti on of the extracted b eam in te nsi ty.

An additional requirement is that the ring magnet

contr ol sy st em be capable of operating i n full rectify;

that is with natural commutation fro m phase to phase.This provides the minimum accelerating time and the

desire d conditions for be am injection into the ZGS.

ZGS Ring Magnet

The ZGS ring magnet circuit contains eight mag -

nets and eight power supplies connected in a se ri es

circuit with four-fold symmetry.

*Work per form ed under the aus pices of the U. S.

The result is the

Atomic Energy Commission.

equivalent of a 12- ph ase supply in each qu ad ra nt of the

ring' magnet.

The rectifiers in each of these eight supplies are

merc ury vapor tubes.

continuously s o that firing control is provided by two

gri ds placed between the cathode ar c and the anode.

The excitation ar c opera tes

Each of the eight supplie s is provided with a low

The f i l -

pas s LC filter. The f i l ter has a rolloff of 40 dB / d ec -

ade and a corn er f reque ncy of about 40 Hz.

te r is underdamped with a damping r atio of about 0 . 3 .

Ripple Amplitudes

Ripple amplitudes produced by the ZGS ring mag-

net power supply were mea sured in two ways.

fir st was to use the f ast gauss clock in the ZGS pro-

grammer .

analog (D-A) con ver ter and to an oscill oscop e. The

gauss pictures fr om the oscilloscope showed that most

of the ripple produced by the original analog controlsys tem was at low frequencies. Amplitudes a t

-

50

cycles (generator frequency) were of the o rd er of 1 or

2 G.

this display.

The

The clock output was sent to a dig i ta l - to-

The 12th harmon ic was not r ead i l y observed on

The second method was to l o o k at the dc voltage

acros s one pair of ring magnets. This was a mor e

convenient signal f o r ripple studies since the relative

amplitudes of the se ver al harmonics were more fa-

vorable for measurement.

The relatively larg e low frequency ripple (a t le ss

than the 12th harmon ic) results f ro m a variety of

causes . One cause is the incorr ect spacing of rectifi-

e r fir ing pulses.

t ics of the low pas s fil ter which attenuates poorly atthe low freque ncie s. Additional cau ses include poss i-

ble e r r o r s in the nu mb er of tu rn s on the polyphase

tra nsfo rme r windings, variations in the leakage in-

ductance from one ph as e winding t o another, lack of

symmetry in circuit resista nces, and imbalances be-

tween phases of the generator voltage.

A second cause is the cha rac teri s-

The effects of some of these ca uses of low fre -

quency ripple can be cancelled on flattop by proper ly

retarding or advancing the firing t ime of one or sev-

e r a l of the 12- ph ase rect i f iers wit h respect to the

others. Fr om this, i t is apparent that exactly uni-

form spacing of the firing pulses to the re cti fie rs will

not produce minimum ripple.

Firing Accuracy

The bias voltages to the unijunctions in the origi-

nal fir ing control s yste m were readjusted to produce asignificant change in the ripple a t the ge nerato r f re -

quency.

sured with the digital phase-angle meter used for

routine monitoring of the ZGS ring magnet supply.

These changes in fir ing angles were m ea-

The data showed that substantial reduc tions in the

low frequency ripple could be obtained if the fir ing

511


2/226

angles of the 12- ph as e r ec t i f i e r s we r e re pr od uc ibl e to

1/10 of an ele ctric al degree.

accu racy and stability of conventional analog ra mp

type control systems. Fo r this reason , a digital con-

tr ol sys tem was designed and installed a t the ZGS.

This is far beyond the

Rectifier Grid Drive

The contr ol of firing an gles to 0.1

electr ical

corresponds to controlling the turn -on of a rec tif ier

to about 5 ps Any variation in the turn -on tim e of a

s ingle rect i f ier m ust be sma l ler than this if the con-

tr ol accu racy is to be achieved.

var iat ion in the average t ime delay f rom one rect i f ier

to another must be sh orter than 5 ps

In a simil ar way,

High voltage switch- t rans is tors w ere used to pro -Thevide the drive t o the grids of the rec tifie r tubes.

fir ing pulse fr om the control ci rcuit was coupled

through a well insulated pulse transformer to a drive

module that operated at rectifi er cathode potential.

The l / 2 p s pu lse wa s lengthened to 4 m s by a single

shot amplifier located at cathode potential.

lengthened pulse was applied to switch- t rans is tors

that supplied the grid drive signal.

This

The grids were switched to t 3 0 0 V with respect

t o the cathode with a rise time of 600 ns.

conditions, the tube turn -on was about 1 ~s a t low

cur ren ts and also at high cu rre nts when the tube was

fi rs t used. At high cur ren ts, the time delay and the

j i t te r increased du ri ng th e f i r s t tw o min ut es. At

equilibrium, the delay varied f ro m a minimum of

about 2 ps to a maximum of about 5 ps We believe

that this increase in turn -on delay is caused by out-

gassing of the electrodes under high current loads.

Under these

Turn-off of the rect ifi er by remov al of the posi -

tive anode voltage left a plasma in th e space be tw een

the grids and the cathode arc. This caused grid cur -

rents that lasted long enough to influence the next

turn-on. For this reason, a switch- t rans is tor was

used to clamp grid#1

to a negative bi as when the tube

was off.electrode space in about 1 / 2 m s .

This bias swept out the ions left in the in ter -

Nu mb er Sys tem and Lo gic

The digital pa rt s of the co ntro l sy st em shown in

the block diagram of Fig. 1 use the binary number

syste m. This was chosen since the input fr om the

ZGS p rog ramm er i s in one 's comp le ment f o r m and

because of co nv en ie nc e in co nst ru cti on .

The DISPLAY sys tem use s LED numeric display

elements driven by a binary coded deci mal counter.

This i s convenient for checkout by the maint enance

personnel.

The digital logic util izes 7400 s e r i e sTTL-MSI

integrated circuits .

chips a re mounted in dual in- l ine packages and are

interconnected by wire-wrap wiring.

delays for this logic ser ie s var y f ro m package to

pack age but a r e of th e o rde r of a few tens of ns.

About 250 integrated circ uit

The propagation

Phase Lock Loop

The generator voltage wave var ie s in frequency

during the ZGS cycle and is very dis tor ted.commutation produces notches in the generator

Rectifier

voltage wave that ar e as lar ge as 25% of the peak volt-

age. Fo r this reason, a zero cross ing type of r e fer -

ence is not suitable even when the input wave i s

heavily fil tered. This is t rue because the f i l ters that

were tr i ed produced variable phase shifts and l o r large

trans ient s when entering flattop and invert. The

phas e- lock loop shown in the upper righ t portion of

Fig. 1 was found to operate satisfactorily.

In this loop, a dc tachometer voltage fro m the

main gene rator and the output voltage f ro m the fil te r

ar e added in an operational amplifier. The resul tant

voltage dri ves a voltage- to- frequency (v-

f) converter

that prod uces a sawtooth wave. This sawtooth dri ves

a binary that divides the frequency by two and pro -

duces a squ are wave with*lO V output levels.

It is well known that the mathematical product of

two sine waves of the sam e frequency but different

phases produces a dc component plu s a se cond h a r -

monic component.

pha se di ff ere nce be tw ee n the two si ne wa ve s.

The dc component depends on the

In a sim il ar way, the product of a squar e wave

and a distor ted sin e wave of the sa me period prod uces

a dc component that is a function of the ph ase differ -

ence between the two waves. This fac t i s utilized in

the loop in F ig . 1 by feeding the disto rted sin e wave

fr om the gen erator bus and the squa re wave f r o m the

frequency divider to a transconductance type analog

multiplier.

po nent s is fed to the low- pass fil ter to co mpl et e th e

feedback loop.

The output which contains dc and ac com-

This loop would operate without the dc tachometer

input; but since the maximum loop gain is finite, the

ph ase dif fer en ce be twee n the a c re ference and th e

squ are wave in locked op eration would va ry with gen-

era tor speed. The dc tachometer input amplitude is

adjusted so that the dc output of the fil ter is near ly

ze ro in the locked condition. In this way, the phase

difference between the ac refer ence and the s quar e

wave is made very near ly 90° and independent of gen-

erator speed.

Main Counter Loop

The MAIN COUNTER and its associated feedback

loop ar e shown in the upper left pa rt of Fig. The

function of this c ounter i s to provide a digita l deg ree

scale that has 360. 0 per generator cy cle an d that i s

phase sy nc hro niz ed wi th th e squa re w av e f r o m the

phase- lock loop.

1.

Its operation is described below.

The dc tachometer voltage, shown in the upper

left of Fig. 1, is multiplied by a constant in the

SERVO CONTROL.

input of the v-f converter.

jus ted to give v e r y nearly 160 co un ts per e lect r ical

deg ree of the gen era tor wave. In this way, the maincounter receives ve ry near ly 16 x 3600 counts eac hgenerator cycle, without the feedback loop's co rre c -

tion. This make s the action of the feedback loop al-

mo st independent of genera tor spee d.

The product is then applied to the

The gain constant is ad -

The main counter is provided with gates that

clea r the counter to zero each time the total countreaches 16 x 3600.

pulse a t the s am e t ime that the counter is cleared.

N o other input cle ars the counter , thus assur ing

360. O o pe r m ain cou nte r cy cl e.

The gates generate a ROLLOVER

5 12


3/226


4/226

If the prog ram mer input is O o (full rectify) and i f

the FEEDBACK is not in us e (F= 15 ), the B input to

the comparator may be fro m 15' to 45O depending on

the D.S. 1 setting.

As soon as the number in the pre set counter

equals or exceeds the number B, a compari son pulse

is generated.

#1 and causes the hCNT to advance to the second

state.

This pulse is transmitted to rectifier

In this second state, D.S. 2 digital switch is con-

nected by the sel ect or to output E and driv er 2 i s

conditioned.

pu t has a va lu e of 30.0'.

In this ca se, the 30°1NC generator out-

This pro ce ss continues to fire each of the 12 rec -

tifi ers in the proper sequence.

synchronizing at star tup of the sys te m becaus e of the

equal to or g rea ter than action of the compa rator .

The ACNT i s se l f -

Feedback

The digital control sy ste m is provided with a fast

channel that ma y be used in cpnjunction with other

control syst ems such as the B feedback cont rol and

the external be am intensity control.

The feedback sys te m, located a t the bottom ofFig. 1, consists of an input ampli fier , a S /H , an a

-

d

con ver ter , and a re gi st er (REG). The input voltage

range is - 20 to t 2 0 V .

pu t ra ng e of 0 to 30. Oo

not in us e. the REG is held at 15. 0 .

The a-d converter has an out-

When the feedback input is

Model

Dynamic checks of the digital control of the recti -

fi er fi ring angles a r e done through use of the MODEL.

This model contains two smal l 6- ph as e t ransformers

supplied f r o m the generator bus.

has a delta primary and the other a wye connected

p r imary .

coupled SCR ' s that a re operated by the 12 dr iv er s of

Fig. 1. The rectified voltage is fed to an operational

ampli fier analog model of the ZGS passi ve fi lte r and

rin g magnet system. Magnet voltage, magnet cur -

rent, and current through the filter inductor ar e

available.

transmitted to the power house display sys tem f o r the

operators' inspection.

One trans former

The secondaries connect to 12 small photo-

Model current and magnet voltage are

Discussion

The control syst em has been in routine use since

Initial start up was with all the digi- No ve mb er 1973.

tal switches, D.S. 1 - D.S. 12, se t for equal int er -vals between rectifier firings. No parti cular prob-

lem s with a r cbacks wer e encountered.

Readjustments of the individual rec tif ier angles

wer e made to minimize ripple. This was done with

the aid of the control computer using the PHASOR

program. The am pl itu de s of the f i r s t , se co nd , and

third harmo nics were reduced to values that were

about a factor of 20 sma lle r than with the old analog

ramp control system.

back in pu t.

2

This was done without the feed -

The amplitude of the si xth har mon ic could be

minimized but remained large.fai rly broad, and the values of D.S. 1 - D.S. 12

The minimum was

indicated that the delta and wye connected transform -

e r s did not produce voltage waves that are 30° apart.

It appea rs that the angles ar e about 28O.

The feedback connection is used wit h the B coi l on

the ring magnet to control the slope of the magne t flat -

top for energy los s extraction. In this mode, in-

crease d ripple amplitudes appear on the ring magnet

voltage.

pr od uc ibl e f r o m cycl e to cy cl e.

frequency noise is picked up in the B sys tem .

so, the ripple is less than with the old control system.

The feedback input is used during resonant ex -

traction to control the extraction r a t e .3 The signal is

supplied fr om beam monitoring devices to magnetical -

ly pro gra m the beam position during extraction.

These a re at low frequencies and ar e not re -

It ?ppea r s that low

Even

Acknowledgments

Many people have enthusiastically contributed to

the digital firing control sys tem project.

forts have made the sys tem operational in a re mar k -

ably short time.

Their ef -

We wish to thank Mr. Ray Kickert fo r his e ffor ts,

especially during the measu rem ent s of the dynamic

charac teris tics of the me rc ur y vapor rectif iers and

during debugging and testing of the system.

The ring magnet power group, under Mr. George

Wes t, did an outstanding job of cons truct ing ma ny of

the circuit modules and making the many changes in

the control and interloc k cir cui ts. We wish to thank

P. Bertucci, L. Johns, P. Roth, E. Kulovitz,W. Welch, and the whole ring magnet power group.

1.

2.

3.

References

J. F. Sel le rs , E . F. Frisby, W. F. Praeg, and

A, T. Vis ser , Ring Magnet Power Sys tem for the

Zer o Gradient Synchrotron, 1965 Par ti cl e Accel-

er at or Conference, Washington, D. C., March 10

to 12, 1965, IEEE Trans actio ns on Nuclear Sci -

ence, Volume NS-12, No. 3, p. 338 (1965)

Lloyd G. Lewis and Anthony D. Valen te, Argon ne

Nati onal La boratory, PHASOR:

pu ter Method for Di sp la ying Ampl it ude and Phase

of Ripple Components in the Ring Magnet Voltage,

IXth International Conference on High Energy

Accelerators, Stanford, California, May 2 to 7 ,

1974.

A Control Com-

Y. Cho, E. A. Crosbie, L. G. Lewis, C. W.

Potts, and L. G. Ratner, Argonne National Labo-

rat ory , Slow Resonance Extract ion of Two Simul -

taneous Beams without R F Structure, IXth Inter -

national Conference on High Energy Accelerators,Stanford, California, May 2 to 7, 1974.

5 14


5/226

v D.

C.

0 D + A REG. LOAD TA CH .D.C. TACH.

v v

SERVO CONTROLSUB.

v-*

f S U M

RECTIFY D.S.160 COUNTS

/

DEG.

COUNTS / DEGREE

START

DISPLAY

CONTROL

I

A D DI

G

I

S E L - f 256

MCOMP ADD

+ + w

SU B. RATE LAG REG.

9

I

R E G . A D D

MAX.I N V E R T

ADJUST

1 I

SUB,

0005'

0005

S E L .

CURRENT IN VE RT MAN,

COMP. PROGRAMMER

F

REG. k 5 O

1CONTROL

TIMING

0

0

0

0

0

0

0

0

30 INC.

S E L E C T O R

D.S.DECOD

- - - - _ -

D.S.I

12

F E E D I N D I V I D U A L D R I V E R S

BACK ANGLE ADJ UST

FIG. 1 - DIGITAL FIRING A N G L E CONTROL SYSTEM BLOCK DIAGRAM

5 1 5

t


6/226

PHASOR: A CONTROL COMPUTER METHOD FOR DISPLAYING AMPLIT UDE

AND PHASE OF RIPPLE COMPONENTS IN THE RING MAGNETVOLTAGE*:

Lloyd G. Lew is and Anthony D. Valente

Argonne National Laboratory

Argonne,

Abstract

Deviations of 12- ph ase generator - t ransformer -

rect i f ier sys tems f ro m ideal performance resul t in

the generation of ring mag net ripple a t the fundamen-

tal and at low harm onic s of the gen era tor frequency.Analysis and display of the amplitude and p hase of

each of these harmonic components makes it po ss ib le

for the sys tem s engineer to minimize the ripple by re -

adjusting the control system. The Zer o Gradi ent Syn-

chro tron (ZGS) control comp uter sampl es both the f i l -

ter ed voltage on the ring magnet and a ref erenc e volt-

age produced by a phase- lock loop connected to the

generator bus.

val of the ZGS cycle and are then analyzed.

driven graphic displays plot the raw ripple data, am -

pl it ud e an d phase bar graphs for each h armoniccorn-

ponent , reconstructed ri pp le da ta for checking, and

graphs fo r comparing ripple components under differ -

ent operating conditions. Numerical information i s

als o displayed. The PHASOR pr og ra m co rre ct s the phase ang le s for th e phase sh if ts pro duc ed by th e ZGS

pass ive fil ter. These correc ted angl es in di ca te whlch

of the 12- ph as e fi r in g angl es should be re ta rded or acl-

vanced to reduce the ripple.

Data ar e taken during a selec ted inter -Computer

Introduction

Power supplies for synchrotron ring magnets

usually consist of phase -controlled rectifiers operated

fro m an ac power l ine or f rom ac generators. The

number of a c phases is often high to minimize the rip -

pl e. Th e ri pp le on the magnet is reduced even more

in some syste ms by inserting a low- pa ss f i l te r be-

tween the rec tif ier s and the ring;nagnet.

In a 12- pha se sys tem with a low- pass fi lt er , the

amplitude of the ri pple component at the fundamental

frequ ency of the power line ma y exceed the amplitude

of the 12th harmonic. This i s caused in pa rt by the

fil ter 's attenuation being la rg er at the higher frequen-

cy. Additional cause s include er ro r s in the number. OE

turns on the polyphase tran sfo rme r windings, variation

in the leakage inductance from one phase winding to

another, lack of symmet ry in circuit resis tance s, and

imbalances between phases of the pow er li ne or ge n-

erator .

The magnitude of the low frequency ripple c om -

po ne nts may be redu ced on fl at to p by proper ly re ta rd -

ing o r advancin g the firi ng ti me of ea ch of the 12 -

phase rect i f iers wi th respec t to th e others.

Pr op er adjustment of the fir ing delays ofcach

of

the 12 phases i s extremely di ff icul t by t r i a l and e r r o r

methods and an optimum i s difficult to determineby

eye. Fo r these reasons, analytical methods ar e re -

quired.

A swept frequency spec trum analy zer was used

::::Work

performed under th e auspices of the U. S.

Atomic Energy Commission.

Illinois

in a n attempt to analyze the ripple during a 700 m s

flattop of the ZGS ring magnet.

quencies, 50-600 Hz, the sweep ra te for the requ ired

resolution was so low that an analysis could not be p er -

for med in the length of the flattop, In addition, no

phase ang le informat ion i s giv en by su ch an instrument .

At the low ripple fre -

The ZGS control computer s yst em was used in

conjunction with the program PHASOR to analyze the

ripple and display the results . The pr og ra m i s of the

interactive type where the computer operator directs

the logic flow after observi ng the re sul ts of each se c -

tion of the program.

Data Input

The reference for al l phase angles was a square

wave produced by a phase- lock loop that had one phase

of the gene rato r voltage a s it s input. This loop acted

as a f il ter for the distortions in the genera tor voltage

wave and gave sh arp indications of corresponding

poi nt s i n each generator vo lt ag e cy cl e. Th e lo op wa s

adjusted so that the positive -going sq uare wave tra nsi -

t ions were at 9 ° on the gene rator voltage s ine wave

and wer e independent of gen erato r voltage and f re-

quency.

This squ are wave, shown in Fig.lA,

was the in-

put to an an al og integrator that was vo lt ag e limited at

* 10 V. The waveform fr om this i s shown in Fig. 1B.

The ri se and fall t imes were adjusted to be slightly

longer than the control computer data sampling i n t p r -

Val. In this way, the computer was as su re d of one

data point on the r l s e even through the compu ter was

not synchronized to the gener ator.

The second input to the ZGS control computer

data station was a voltage cGntaiiiing the magnet r ipp l r

rnformation. It was obtained by capaci tively coupling

the voltage which was a cr os s one quadrant of the ring

magnet to an amplifier with a high common mode r e -

je ct io n. Zenerdlode

networks el iminated most of the

C ~ L component.

Data Taking

The manual keyboard at the computer driven

scope was used to specify the point in the ZGS cycle

lor the st ar t of data taking (e. g. 250 ms af ter the s tar t

of flattop). The keyboard was then used to ent er the

niimber

of data sa mpl es of the ripple voltage that ar e

to be taken.

fcr al l of the da ta points.

Z O O

u s intervals and alternate between the reference

wave: and the ripple wave.

The computer then gene rates requisitions

Measurements a re made at

When the op era tor pushes e DISPLAY NEW

SET key, a new set of d ata is ta ken on the next ZGS

cycle.

screen. Figure 3 top plots the sa mple s taken on the

r e l e r ence

wave.

risin g and falling slopes of the refere nce. Fig1n-e3

bot tom plo ts the samples take n on the ripple wa ve.

These data ar e then displayed onthe

scope

The points that a re circl ed a re on the

516


7/226


8/226

a

a

ALPHA SET0 0 0

0 0 0

0 0 0

0 0 0

0

AMP.- 2I . I

I . I.2

.3

. II .4

. I

.o

.o

.o

.o

6

PHASE0

29.3- I

11.7

I

12.0

-130. I-145.9- 76.6

- 82.1138.0

1 1 1

47.8

- 5 .915 6

Fi g 3.

dcI234

56

789

I O

I I

12

AMP.0000000

000000

RETAINED SETPHASE ALPHA SET

0 0 00

0 0 000 0 00

0 0 0

00 00

0

00

Display New Set and Deri ved Ripple

PHASOR

15 NOV. 1973

21 :04 :36

START AT 16 400.0

300 SAMPLES TAKEN

2 PERIOD ANALYSIS

H Z 1 ST THETA

I 50.15 4.82

2 50.10 2.28

Three hundred data sample s were taken and two p e r i -

ods we re analyzed. The first peri od had a fr equ en cy

of 50.15 Hz, while the second had 50.10 Hz. This dif -

fer enc e is caused by the slowing down of the ge ner ato r

during the ZGS cycle. The angle8d for the f i rs t data

po in t in each cyc le wa s 4.82O and 2 . 2 7O .

The top graph gives the recon struct ed r ipple

The FIT

data that was calculated fr om the amplitudes and

phases of th e derived ri pp le co mp on en ts .

number is also given.

4 , as wel l as in Fig. 3, for identification purp oses and

because ma ny people get a be tter "feel" for the data

from an analog type display.

This graph is included in Fig.

The upper bar graph of Fig. 4 plots the dc com-

po ne nt and th e amp li tu de s of th e f i r s t 12 harmonics of

the ge nerato r frequency.

plo t is shown above and to the le ft of the b a r graph.

The sc ale factor in volts pe r inch is changeable at will

through the scope keyboard. These amplitudes ar e for

the filtered voltage applied to the ring magnet.

The vertical scale for thi s

The lower bar graph of Fig. 4 plots the phaseangles x for the fi rs t 12 harmonics. The range of

angles isrfixed a t t 180° to -180° in a ccordance with

the restriction on x r in equation (8).

have been corrected for the phase shifts produced by

the passive filter of Fig. 2. They are , the refore, the

phase an gle s of th e harmonic co mp on en ts at the outputof the rect ifie rs. This i s done to make it poss ible to

deter mine which rectifie r f ir ing angles should be r e -

tarded o r advanced to reduce the amplitudes of the rip -

ple co mp on en ts .

These angles

The lower part of Fig. 4 lists sev era l set s of

numer ic data. The column in the center l ists dc and

phases 1-12. The two columns to the left of center

give the amplitudes in volts and the phase angles in

degrees for the 12 phases and the am pl itu de of th e dc

component. These a re the values used to calculate

the deriv ed rippl e plotted a t the top of Fig. 4.

The table of numbers at the lower left re cor ds

the alpha angle setting s for the 12 pha ses and in addi-

tion, the full rectify setting. The se ar e the settin gs

of the ZCS control system that produced the ripple r e -

const ructe d a t the top of Fig. 4. The numbers ar e in

octal code and represe nt deviations f rom the ideal uni-

fo rm spacing of rec tifi er firing angles. These num -

b e r s a r e recorded in th e exact format in wh ich th ey

appea r on the con trol panel of the digital firing angle

control system.

These numb ers ar e ente red through the use of

theENTER/MODIFY

ALPHA SET subroutine, and

can be used by the o perat or when desired.

Retained Set

The opera tor may save a se t of re sul ts , such as

that presented in Fig. 4, by pushing the button labelled

RETAIN THIS PHASOR SET. This activates a sub-

routine that stor es the information in computer me mo-

ry f or c omparison with future data.

The ope rat or may then analyze a new se t of data

and call for GRAPH PHASOR SET. The re su lt i s the

gener ation of a display such as shown in Fig. 5 .

518


9/226

1.00

AMP. MAG.

P. s.

I

I I

PHASOR

15 NOV. 1973

21 :04: 36

START AT16

400.0

300 SAMPLES TAKEN

2 PERIOD ANALYSIS

HZ 1 ST THETA

I 50.15 4.82

2 50.10 2.28

RETAINED SET-180

ALPHA SET AMP. PHASE AMP. PHASE ALPHA SET

0 0 0 0n

077

37

76

36

472

36

76

36

74

77

36

77

36

- . 2

I I

I . I

- 2

3

. I

1.4

. I

o

o

o

o

. 6

029.3

I 11.7

- I 12.0- I 30.-145. 9- 76. 6

-82.1138.01 1 . 1

47. 8- 5. 9

15 .6

dcI

3456

789IO

I 1

12

F i g . 4 Graph

The numeri c informati on at the top right of Fig.

5 i s s imi lar to that of F ig . 4 except i t i s for the new

s e t of data.

number are aIso for the new set of data.

The derived ripple curve and the FIT

The ba r graph for the amplitude now has t wo

l ines a t each harmonic locstion.

pa i r gi ve s th e am pl it ud e of th at harmonic co mponen t

in the new set of data, while the r ight l ine in each pair

gives the amplitude of that harmonic component in the

reta ined set . In this way, the amplitudes may be com -

pa r ed visually t o determine the effect of the changes

in the f ir ing angles.

The lef t l ine in each

The bar graph for the phases also has a pair of

l ines a t each harmonic location. The left l ine in each

p a i r gives th e phase an gl e for that harmonic compo-

nent in the new se t of data, while the righ t line gives

the p h ase angle of that harmonic component in the re-

tained set .

The num eri c information, at the bottom of F i g . 5

on the r i ght, gives the amplitude and phase angles fo r

the ret ained se t and in addit ion the alpha s et that pro -

duced them, The numeric information,at

the bottomon the le f t , gives the amplitude and phase angles for

the new set of data and the corresponding alpha set.

output

The p r ogr am PHASOR is an in teract ive one so

that most of the output is through visual observation of

the computer driven scope display. Seve ral options in

the interac tion can produce copies of the scope display

on8%

x11

ph ot og ra ph ic paper. The se include:

V

00

000

000

000

PHASOR Set

0

0

0

0

0

0

COPY NEW SET - - similar to Fig. 3,COPY LEFT SHIFTED - - that portion of F i g . 3that was analyzed,

COPY DERIVED - - derived ripple,COPY PHASOR iALPHA SET - - simi lar toF i g . 4, 5.

The 'hard -copy" unit ca n produce a pr int in about ten

seconds. No l ine pri nter output is provided.

Calibration

The effects produced by changes in rect if ie r f ir -

ing angle were experimentally investigated. Fo r ex-

ample, the firing angle of phase nu mber 1 was ad -

vanced s everal e le ct r ical degrees to produce a larg e

ripple on flattop.

the amplitude and phase of the ha rmo nic c omponents

of the ripple thus produced.

peat ed for seve ra l co mb in at io ns of cha ng e s in f ir ing

angles of selected rectif iers.

PHASOR was then used to meas ure

This procedur e was r e -

The calibration data were useful i n predicting

which of the 12 rect ifie r groups should have the ir fir -

ing angles altered to reduce a given obse rved r ipple.

Discus sion

Analyses of r ipple data wer e made fo r a variety

of firing angle combinations during the calibration runs

and for many actual operating conditions while ripple

reduction adjustments wer e made. In all cases, the

ba r graphs s imi lar to Figs. 4 and 5 showed ve ry s mall

or no amplitudes fo r t he5th, 7th

8th

9th, IOth,

and

1 l th harmonic s. The phase angles computed fo r these

5 19


10/226

PHASOR

15 NOV. 1973

23:25 : 16

START AT 16 600 O

500 SAMPLES TAKEN

2 PERIOD ANALYSIS

AM

HZ ST

THETA

I48. 82 4 58

P.

s. 2 48. 76 3. 27

-180 RETAINED SET ALPHA SET AMP. PHASE AMP. PHASE ALPHA SET

77 32 77

32 76 32

101 32 77

32 76 32

473

-- 6

3

3

.2

.2

.o

. I

.o

o

o

o

o

, 2

026. 4170.8

- 96. 2- 117.8

- 146.0- 78. 3- 63. 7- 166.4- 30.- 68. 0

78. I, 42. 0

dcI23456

7891

I 1

12

\ o

. 3

. I

. I

. I

. I

2. 0

I

. I

.o

o

o

. 7

0- 106.3

43. I- 153.2- 158.2- 73. 6- 77.1

-90. 2

- 153.7-131. 3- 136.6136.0

I38 3

77

46

76

46

470

50

76

47

74

77

47

77

46

F i g . 5 Graph PHASOR Set With Retained Set for Comparison

har moni c components varied widely fro m one data s et

to the next which leads us to the conslusion that these

harmonic components a r e largely the re sult of "noise"

or inaccuracies in the input data.

We we re not able to find a combination of rec ti -

f ier f ir ing angles that produced large or significant

amounts of 5th, 7th, 8th, 9th, loth, and I l th, harmon -

ics. We therefore have amplitude and phase mea -a u r e m e n t s at harmonic numbers l , 2, 3, 4, 6, and 12

for a total of 12 measurements .

The control s ys te m has adjust ments fo r each of

the 12 phases bu t only 11 of these a r e indepe nden t v a r -iables as fa r as r ipple i s concerned. The 12th phase

control and the "full rectify" control adjust the slope

of the f lat top. This slope is adjusted to z ero before

the r ipple measu reme nts ar e made.

that would compute new firing angle settings fr om theripple component amplitudes and phas e angles. Ther e

appear to be enough measur ements to per mit the solv-

ing of a se t of 12 equations.

for this was developed in the ver y l imited effort ex- pended .

ple may be re du ce d one co mpon en t a t a ti me .

the procedu re used.

Some thought was given to writing a program

No ad eq ua te algori thm

Inspection of the problem indicates that the r ip -This i s

A per fect recti f ier sys tem will pr od uc e on ly th e

12th harmoni c, ther efor e this cannot be elimina ted oreffectiv ely reduce d by adjusting firi ng angles. On the

other hand, the 6th harmoni c can be incr eas ed o r de -

cre as ed in only one way. That is, all even number ed

phases should be adva nced an eq ua l amo unt and a ll qddnumbered phases retarded by the same amount. This

method reduced the 6th harmoni c amplitude to a mini -mu m but would not make it vanish. The data indicate

that the phase shift between our delta and wye conne ct-

ed transformers is only 28' rather than the theoretical

30°

The fir st harmonic amplitude may be reduc ed by

changing al l of the 12 angles. In thi s case, the

changes a re dis tributed sinusoidally with the peak of

the distribution determined by the phase angle of the1s t harmonic of the ripple. The 2nd ha rmo nic may be

reduced by a si milar procedure except that the sinus -

oid for the distrib ution is the 2nd harmoni c.

This progr am was used successfully to reduce

the flattop ripple a t the ZGS with the distr ibutions foralpha angle changes determined manually. It is hoped

that additional subrou tines can soon be added to pe r -

mi t the compu ter to calculate the alpha sett ings that

will minimize the ripple.

5 20


11/226


12/226


13/226

LINK TO SDS925 IN MAIN CONTROL

DECTAPES POP-3

KLYSTRON

MAINTENANCE

GROUP

CCR CONTROLS

CCR ANALOGS

LINK TO I B M 360/370 COMPLEX

KLYSTRON GALLERY

CONTROLTO

PULSED PHASECLOSURE,

GUN MOD A

GUN MOD B

CONTROL TO

PULSEDENERFY

VERNIER

Fig. 2

5 2 3


14/226

THE CPS IMPROVEMENTS 1965-1973

AN ASSESSMENT

THE CPS

CERN, Geneva,

Summary

I n 1965, p lans were made t o in cre as e t he beam

in te ns it y delive red by the CPS by a f ac t o r o f t e n o r

more. The f i r s t s ta ge, involving a new power supply fo r

the main magnet and more than doubli ng th e c ycle repe-

t i t i on r a t e , was completedi n

1968. In the second

stage, which i s now es se nt ia l l y complete, the major

i tems was t he cons tru cti on of an 800 MeV slow- cycling

b oos t er i n j e c t o r . Many o t h e r mo di f i ca t i ons we re i nc l u-

ded. The Linac cu rre nt had to be incre ased by an o rder

of magnitude t o suppl y the Bo oste r, and th e highe r beam

in te ns i t ie s r eq u i red a more powerfu l RF accele ra t i ng

system. Besides the 800 MeV inj ect ion e lements , quadrupo l e l e n s e s were i n s t a l l e d t o avo id l o n g i t u d i n a l d i l u-

t i o n a t t r a n s i t i o n , andm u l t i p o l e s

t o co u n t e r ac t i n s t a-

b i l i t i e s . I n ad d i t i o n , t h e chamber vacuum was improve d

by a f a c t o r of t en , s h i e l d i n g and r ad i a t i o n r e s i s t an c e

incr ease d where necessary, and beam-equipment inter -

act ion r educed . Adequate ins t rumenta t ion and con t ro l

f a c i l i t i e s h ad t o b e pr o vi d ed , and t h e e f f i c i e n cy of

fa s t and s low ex t r a c t i on sys tems improved . Per tu rbat ionsdue to v ar i ous co ll ec t i ve phenomena had t o beovercome.

The performance ob ta ined dur ing the f i r s t phys ics

runs i s r epor ted .

1. I n t r o d u c t i o n

Af te r a few yea rs of op era ti on, th e CPS had reached

a maximum intensity of 1Tp/pulse*

and,i n

view of space-

charge ef f ec ts , a fu r t her f a c to r o f two seemed the mst

th at could be expected. The motor -genera to r s e t supp ly-

ing th e main magnet l imi te d t he du ty cycle to a t yp ic a l

value of 10%a t 19 GeV/c (200 ms f l a t- to p with a 2 sr e pe t i t i on t ime) and even less a t h i g h e r e n e r g i es (100

m s f l a t- top every 3 s a t 24 GeV/c). The exper imental

f a c i l i t i e s co mp ri sed two h a l l s , w i t h a t o t a l a r ea of4000 m2, f ed by i n t e r n a l t a r g e t s and a s i n g l e f a s t

ex t r ac t i o n ch an n e l .

I n 1964, an improvement programme w a s launched withthe ob j ect o f increas i ng t he average acc ele ra te d beam

i n t en s i t y by a f ac t o r of 10 t o 15'. This was to be

achievedi n

two stages :

i )

i i )

r a i s i n g t h e r e p e t i t i o n r a t e t o g a i n a f a c t o r o f 2

o r 3 , depending on energy and flattop leng th , by

constructing a new magnet power supply;

r a i s i n g t h e i n j ec t i o n ene rg y ( f ac t o r 5 i n in ten-

s i t y per pul se) . Two possi ble methods were inves-

t i g a t e di n

d e t a i l ; a 200 MeVLinac'

and a 600 MeV

twin s low cycl ing boos ter s ynchro t ron 3 , A compara-t i v e s tudy4 showed t ha t alth ough bo th schemes could

pro du ce t h e r equ i red i n t e n s i t y i n c r ea s e , t h e h igher

space- charge l i m i t of t he boos ter a l lowed a g reater

p o t e n t i a l f o r f u t u r e de ve lo pm en t. Fu r t he r s t u d i e s

f i n a l l y l e d t o a n 800 MeV boost er wit h 4 super -

po se dr ings ' .

I n ad d i t i o n t o t h e new i n j e c t o r , t h i s

p a r t of t h e programme inv ol ved a number of comple-

mentary improvements t o t he 50 MeV Linac and t he

main p ro ton synchro t ron , which ar e deta i le dbelow.

* Tp = 10 p rotons (Terapro ton) .

STAFF

Switzer land

'The programme was balanced by a comparable expan-

s ion of exper imenta l ar eas and f a c i l i t i es (West Hal l

wi th t h eOmega

spec trom ete r and t he Big European Bubble

Chamber(BEBC)

a nd n e u t r i n o f a c i l i t y w i thGargamel le) ,

which took place simultaneously.

2 . Main Magnet Power Supply

The new power supply6 was designed to more tha n

double the du ty cycle .

The magnet vo lt ag e was incr ea sed from 5.4 to 108

V,

which approx imate ly halved t he r i s e and f a l l t imes

of t h e mag ne t ic f i e l d . I n o r d e r t o av o i d i n c r eas in g

the maximum vol ta ge t o ground, l im ite d by t he winding

i n s u l a t i o n , a se co nd r e c t i f i e r s e t was i n s e r t e d i n t h e

middle of the magnet c i r c u i t , w i th the ou tpu t vo l tages

of bo th r ec t i f i er se t s symmetr ica l to g round . Keep ing

the same maximum current (6400 A) as b e f o r e , t h e h i gh e r

magnet voltage implies a higher peak power, namely

95 MVAi n

p l a c e of 46 MVA.

Th e i n c r eas e i n d ut y cy c l e r a i s e s t h e av e rag e

power and t h e l o s s e s i n t h e magnet . Mean power r o s e

from 18 to 46 MVA and power dissipation i n the magnet

f rom 1.6 to 3 MW. The magnet cool ing system had t o be

adapted to the new cond it io ns.

The new power sup ply has a more fl ex i bl e co nt ro l

system, which provides a wider cho ice of magnet cyc les ,

i n c l u d i n g t h e p o s s i b i l i t y o f two " f l a t- tops" a t d i ff e-

r en t en e r g i e s . T h e d i s t r i b u t i o n of acce l e r a t ed pr o t on s

be tw een u s e r s i s thereby simplified; a common example

of such a complex cy cl e i s : acce l e r a t i o n t o 2 6 .3 GeV/c,e j ec t i o n o f 4 bunches t o ISR , d ece l e r a t i o n t o 2 4 GeV/c,

t h en sl ow ex t r ac t i o n s h a r ed w it h an i n t e r n a l t a r g e t o v e r

a 400 ms b u r s t .

The reduction obtainedi n

the r ipp le vo l tage (20 V

peak t o peak i ns t ead of 100) and the bet t er r eproduci-

b i l i t y of t h e magnet f i e l d (4 a r e important fac-

to r s i n p roducing a sa t i s f ac t o ry s low ex t r ac te d beam.

Re li ab il i t y has proved t o be very good (2 h down-

t ime per 1000 hours o f op era t ion i n 1973).

An important addi t io nal imp li cat ion was the need

to in crea se the mean power and the r a te o f r i s e of the

aux i l i ary power sup pl ie s . These modi f ica t i ons were

car r ie d ou t p rogress ively , and s t i l l cont inue today , as

each aux i l iary sub-system proves t o be a bott le- neck

f o r an i n c r eas e i n t h e m achi ne o v e r a l l e f f i c i en c y and

h as , i n i t s tur n, to be matched t o t he main power sup ply

cap ab i l i t y o r modi fied to improve th e co n t r o l o f beamdynamics ef f ec ts .

3. Linac

Since the o r i g i na l 50 MeV Linac had a l so t o serve

as i n j ec to r fo r the new Boos ter synchro t ron , i t s per -

formance require d su bs ta nt ia l improvement. This involved

i n c r eas i n g t h e p u l s e l en g t h t o 100 u s , f o rm u l t i t u r n

i n j ec t i o n up t o 1 5 t u r n s ; m ore cu r r en t (100 mA w i t h i n

a spe ci f i ed em i t tance and energy spread (30 I mm mrad

5 2 4


15/226

and 2 150keV);

and a h i g h e r r e p e t i t i o n ra te ( 2 s - ’so t h a t a l t e r n a t e p u l s e s c ou ld b e sent down a p a i r ofnew beam measur ing l in es . Besides i ncre asi ng t he duty

cyc le of sev era l components ( i on source,pre-acceleratn ,

pu lsed qua dr up ole s , e t c . ) , a major problem was ca vi ty

beam l oad ing and i t s compensation. This was ta ckle d by

i n s t a l l i n g f o r e a c h of t h e t h r e e t a n k s an a d d i t i o n a lRF am pl if ie r usin g more powerful tubes . However, t h e

beam l oad ing co mp en sa ti on i s v er y d i f f i c u l t t o a d j u s t

w i t h a de q ua t e s t a b i l i t y f o r l o ng p u l s e s a t pe ak i n t e n-

s i t y , a n di t

has been necessary t o l i m i t the beam t o

50mA,

toach i eve s t a b l e opera t i on and r ep roduci bl e

beam q u a l i t y .

4 . 800 MeV In j ec t i o n Syst em

A f t e r i n j e c t i o n a t 50 MeV from the L inac and acce-l e r a t i o n i n t h e B o o st e r , t h e 800 MeV beam i s i n j e c t e di n t o t h e PS over one turn , and the bunches ar e t rapped

d i r e c t l y i n synchron i zed bucket s . The i n j ec t i o n sys t em

t oget her wi t h t he assoc i a t ed beam observa t i on dev i ces

and low energy magnet ic corr ec t io ns i s descr i bed else-where

7. An incoming beam wit hin t he sp ec if ie d charac-

teris tics is t r ap p ed w i t h b a r e l y d e t e c t a b l e l o s s e s .

5. Accelerat ing System

The r educt i o n of t he magnet ic f i e l d r ise t i m e a l s o

i mpl i es an i ncre ase o f t he energy ga i n per t u rn , andi t was i n i t i a l l y in te nd ed t o a c h ie v e t h i s w it h a s e t of

t h ree add i t i ona l nar row- band se co nd -harmonic c avi t ies’ .

These would have been switched on 80m s

a f t e r i n j e c t i o n

when a remaining frequency swing of only 10%was needed

t o r each t op energy. Al though p ro t o t ype un i t s were deve-l oped and suc ces s fu l l y t es t ed wi t h t he beam, t he p ro-

j e c t was dr oppe d when, dur ing the second s t ag e of t he

improvement programme, i t became c l e a r t h a t t he wh ol e

RF sys t em wou ld have t o be r e bu i l t .

An add i t i ona l r equ i r emen t w a s t h e d e s i r e t o b e

a b l e t o t r a p t h e 20 Booster bunches i n 10 of the PS

bucke ts a s a means of increasing the ISR l umi nos i t y .

A f t e r i n v e s t i g a t i n g se v e r a l a l t e r n a t i v e sg

, i t was de-

c i d e d t o b u i l d a new accelerat ion system capable of

c op in g b ot h w i t h t h e f a s t e r r a t e o f rise br oug ht abo ut by t h e new magnet power su pp ly ( s e c t i o n 2) and the

h i gher beam i n t ens i t y .

The new RFsystemio

compr i ses t en un i t s spaced

around theVing.

Each un i t has two i de n t i ca lf e r r i t e -

tuned cav i ty res onat ors , working over the f requency

range 2 . 5 - 10 MHz, p rovi di ng a pe ak a c c e l e r a t i n g vo l-tage of 2 x 10 kV. The av ai l abl e power output i s 90 kW

pe r u n i t , wh ic h i s adequate, under the worst condi t ions ,

f o r a n i n t e n s i t y of 1 . 5 Tp per PS bucket . The p a i r ofr e s o n a t o r s i s connected

i n

p a r a l l e l , which s i m p l i f i e s

t un i ng cu r r en t co n t ro l and a l l ows a l a rg er t o l e rance

fo r t he power t ube ou t pu t capac i t ance . The acc e l er a t i n g

gaps are s h o r t- ci r cu i te d by vacuum re lay s a t the endof t he acce l e ra t i on phase of t he cyc l e , so t h a t t h e yshow a low impedance t o t he beam and re- bunc hi ng i s

avoided.

The power amplifier i s a n e u t r a l i z e d 70 kW t e t r o d e ,

operat ing wi th grounded cathodei n

c l a s s B . It i shoused

i n

t he c av i t y compart ment t o p rovi de i s o l a t i o n

be tw ee n t h e vary ing c a v i t y im pe da nc e and t h e f eed ca bl e,

andi s . b u i l t

as a p lu g- in assembly fo r r apid exchange;

t h e rest of the system i s i n t h e c e n t r e of t h e r i n g

where i t i s always accessible. A l l su b- assemblies are

e a s i l y i n t e r ch a n g ea b l e a n d, a p a r t f ro m t h e f i n a l st a g e s,

f u l l y t r a n s i s to r i z e d .

The beam con t ro l system has a l s o been replaced to

meet t he more s t r i nge n t opera t i ona l r equ i r emen t s ( au t o-

matic ph as e programme , ada pt ed pi ck -u p s e n s i t i v i t y ,

beam-derived freq uency programme, sync hron iza tion with

t he Boos t er , s i n g l e bunch acc e l er a t i on ) .

6 . Vacuum System

It has long been known th at re sid ual gas could be

a sou rce o f beam i ns t a b i l i t i es , and t herefo re set alower l i m i t on ul t imate performance than s imple gas

sc a t t e r i ng e f f e c t s wou ld i nd i ca t e . Fu rt hermore , t he

p rospe c t of i nc re a s e d r a d i a t i o n dam age , an d t he re f o r ereduced r e l i a b i l i t y o f vacuum se a l s made o f o rga n i c

m a t e r i a l s , was a n a d d i t i o n a l r e a s on f o r r e d e s i gn i n g t h evacuum sys

eml

.

The 8 2 o i l d i f fu si on pump groups have been replaced

by abo ut 130 s pu t t e r - ion pumps (200 o r 400 2 1s pumping

speed according t o the local load) and 14 turbomolecu-l a r pump groups ( 260 k / s ) f o r pumping down to th e l od5Torr range. A l l the rubber se al s have been replac ed by

metall ic typ es , and new bel lows- s e al e d v a l v es i n s t a l l e d .

The completionof

t h i s p ro j e c t has r educed t he mean

p re s s u r e by a f a c t o r of t e n , nam ely fr om 2-3 lob6 Torr

down to 2-3loe7

Torr. Recent beam dynamicsexperiment&’

have shown tha t a t t h e i n t e n s i t y l e v e l o f 2 Tp/p a

re t u rn t o t he o l d p res su re l eve l i mmed ia t el y lowered t hei n t e n s i t y by 50%. It shou l d a l so be no ted t h a t , i n s p i t eof t he l ong er pump down time, th e t i m e l o s t du e t ovacuum system f a u l t s has gone down from 20 h t o 10 h per

1000 h o f opera t i on .

7 . Extract ion Systems

Ef f i c i en t shar i ng o f acce l e ra t ed p ro t ons be tween

an inc re as in g number of us er s demanded th e development

of new ex tr ac ti on systems and components. Li mit ati on of

t h e i n t e n s i t y p e r m i s s i b l e on i n t e r n a l t a rg e t s , t o avoi d

bot h overhea t i ng of the t a r g e t head andr a d i a t i c n

damage

to adjacent components , p laces a premium upon metho ds

of s low ext ra ct i on which can s imultaneously sha re th e

beam wi thou t un du ly i n c re a s i n g l o s s e s . A resonant ext rac-

t i o n sys tem o f t h i s k i nd i s now i n o p era t io n 1 3 . Gene-

r a l l y , t h e u s e o f h i g h e r i n t e n s i t i e s i m p li e s t h e n ec es-

s i t y fo r i mprovement s i n ex t r ac t i on e f f i c i e ncy .

This problem has been t ack l ed i n two ways; f i r s t l y ,

by t h e de ve lopm en t of e x t r a c t i o n sys te m compone nts wi th

wider ape r tur es f or the same def l ec t in g power; secondly ,

by t h e u se o f dev ices ah ea d of t he e x t r a c t o r magnet

which enhance the se par at i on of the protons- to- be- ejected

wh i l s t providi ng the minimum obst r uct ion i n th e machine

ape r t u re ( sep t a) . B r i e f de scr i p t i on s o f t h ese component s

fo l l ow.

i A Fu l l Aper t u re Ki cker FAK), t o r e p l a c e t h e p l un-

g i ng par t i a l aper t u re dev i ces wh i ch cou l d on l y

ha ndl e beam of pr e- boos ter di me ns io ns .

The new system

si on l i ne magnet modules of 15 Ohm c h a r a c t e r i s t i cimpedance. With a pu l s e v o l t a g e of 40 kV 80 kV

on t h e p u l s e g e n e r a t o r ) , t h e f l u x d e n si t y i n t h e

53 mm gap i s 630 Gauss and t he t o t a l k i ck s t r e ng t ha t 26 GeV/c g i v e s a displacement of 19

nun

a t t h e

septum ext ractor magnet l oca t i on wi t h a 55 ns (10t o 90%) r i se t i m e . These parameters have been

chosen t ak i ng i n t o cons i dera t i on t he expec t ed lar -

ger t rans vers e emi t ta nce and the longe r bunch

leng th of the h igh in te ns i t y beam. The system wascommissioned i n 1973 and has performed w e l l up to

c o n s i s t s of 9 f e r r i t e transmis-

5 2 5


16/226


17/226


18/226

18. Keyser R.L., "The development of la rg e aper tu re

septum magnets fo r slow e j e c t i o n" , CERN i n t e r n a l

r e p o r t MPS/SI/Int. MAE 72-5.

19. B e r t o l o t t o R . , "P r o p o si t i o n p ou r l a c o n s t r u c t i o n

de nouveaux aimants septum pour l ' l j e c t i o n r a pi -

de", CERN i n t e r n a l n o t e MPSfSRINote 72-4.

20. Hardt W . , "Gamma-transition- jump scheme of the

CPS , Proc. of th is Conference .

21. -Gareyte J ., Sacherer F ., "Head - t a i l t ype ins a -b i l i t i e s i n t he CERN PS and Booster " , Proc. of

this Conference.

t i e s - Thaory", Proc. of t hi s Conference.

l i n e a r s p a c e- charge for ces and octup oles" , Proc.

of th is Conference.

-Sacherer F., "Transverse bunched beam in st ab il i -

-Mghl D . , S c h h a u e r H . , "Landau damping by non-

22. Gyr M . , "Studies on the PS o c tu p o les , CERN i n t e r -

na l note MPS/SR/Note 71-16/Corr.

23. Boussard D . , Gareyte J . , "Damping of th e lo ng it u-

d i n a l i n s t a b i l i t i e s i n t h e CERN PS , Proc. of

8t h In t . Conf. on High Energy Accel. , CERB, 1971,

pp . 317-320.

24. -M6hl D . , "Equipment responsible for t ransverse beam i n s t a b i l i t y i n t h e PS", CERN i n t e r n a l n o t e

MPS/DL/Note 74-6.

- F a l t e n s H, U m s t i t t e r H . H . , "Longi tudinal coupl ings

impedances at insulated PS vacuum chamber flanges" ,

CERN i n t e r n a l n o t e MPS/LIN/Note 74-5.

25.

2 6 .

27.

28.

29.

30.

31 .

A g o r i t s a s A. e t a l , 'Techniques fo r measur ing beam

parameters" , Proc. of 2nd USSR Nat ional Co nf . on

Part. Accel., Moscow, 1970.

Carpenter B. , "Experiments with interact ive con-

t r o l s o ft w ar e a t t h e CERN PS", Proc. of I E E Conf.

on Software for Contro l , Warwick, England, 1973.

Madsen J . H . B . , "The expan sion of t he PS c o n t r o l

system", CERN i n t e r n a l n o t e MPS/CO/Note 72-42.

Gouiran R . , "La r a d i o a c t i v i t s de l ' a i m a n t du CPS

e t s o n i n fl u e n c e s u r l a maintenance de l 'anneau -S t a t i s t i q u e s e t p r l v i s i o n s , CERN i n t e r n a l r e p o r t

CERN/MPS/SR

73-5.

Coet P . , "The work i n h i g h l y r a d i o a c t i v e e x p e r i-

mental are as of t he PS", CERN i n t e r n a l r e p o r t

CERN/MPS/MU-EPf 72-2.

Ste in b ach Ch., "Beam dumping, the s i t ua t i on i n

January 1974 , CERN i n t e r n a l n o t e MPS/OP/Note

74-4.

Hoffmann L., " E as t h a l l t r a n s fo r m a ti o n p r o j e c t" ,CERN i n t e r n a l n o t e MPS/MU/Note EPf73-12IRev.

5 2 8


19/226

SIMULTANEOUS STEERING OF H N D

H- BEAMS AT LAMPF*

by

K. R. Crandall and W. E . J u l e

Uni ver s i t y of Cal i f o rn i a

Los Alamos S ci en t i f i c LaboratoryLos Alamos,

Summary

Based on an aly sis and computer s imul at io ns, i t has become apparen t t h a t two k inds of doubl e t s t e e r i n g aren e c e s sa r y f o r l i n e a r a c c e l e r a t i o n of

H

andH

beams

s i mul t aneous l y . The r esp ec t i ve advan tages o f s t eer i ng by e l e c t r i c a l l y " t i l t i n g" and "d i sp l ac i ng" a system ofd o u b l e t s are d e s c r i b e d . S i n g l e d i p o l e s t e e r i n g f o r si-

multaneous beams and the e ff ec t of t he e ar th ' s magnet icf ield on opposi tely charged beams is a l so cons i dered .

F i n a l l y , t h e i mp lement a ti on of this k i nd of s t e e r i n g a t

LAMPF is d i scussed .

Theory

Conven t i ona l s t eer i n g superposes a d i po l e f i e l d t o

c o u n t e r a c t any mechanical misalignment of a quadrupole.

I f o n e is deal i ng wi t h a s i n g l e qu a dr u po l e, t h i s steer -i n g f i e l d c a n b e a p pl i e d i n a ma nner w hi ch e l e c t r i c a l l y

moves th e magnetic cen ter of the quadrupole f rom i tsm e c ha n ic a l c e n t e r t o t h e c o r r e c t de s i g n p o s i t i o n . I n

F i g . 1 a r e shown for ce versu s d isplacement d iagrams fo r

focu ssin g and defocussi ng quads.a d i s t ance , x , f rom t he quad cen t er woul d f e e l a f o r c e ,f,, as def i ned by t he so l i d l i n e .

superposed on t h e qua d f i e l d s im p ly t r a n s l a t e s t h e

so l i d l i ne t o t he dashed l i ne , and moves t he magneti cc e n t e r of t h e quad t o po i n t B . This type of d isp lace-ment works fo r beams of opp osi te cha rge, s in ce , whi le

a quadrupole changes from focussing F) t o d e f o cu s s i ng

(D) when th e si gn of th e beam changes, the s t ee r i ng

f i e l d a l s o r e v e r s e s d i r e c t i o n ,more complicated when one considers two quadrupoles

pos i t ioned c l o s e toge t her which act as a doublet . I n

the s tandard approach, a series s t e e ri n g c o i l i s woundon the two quadrupoles and the same stee r in g f ie ld isap pli ed t o both magnets. From th e above argument on

how t o e l e c t r i c a l l y d i s p l a c e a quadrupole ' s magnet icc e n t e r , o n e m i gh t c on c lu d e t h a t e q u a l f i e l d s a p p l i e d i nt h i s way wou ld e l e c t r i ca l l y d i sp l ace t he doub l e t (bo t h

magnets are housed i n t he same c a s e ) . F i g . 2 showst ha t t h i s t ype of s t ee r i ng does no t p roduce t he des i r ede f f e c t .

A p a r t i c l e d i s p l a c e d

A s t e e r i n g f i e l d

The s i t ua t i on becomes

D i p ol e s t e e r i n g f i e l d s a p p l ie d i n t h i s way w i l le l e c t r i c a l l y "tilt" t h e d o u b le t , n o t d i s p l a c e i t .

kind of s te er in g, which w e w i l l c a l l " t i l t e d d o ub le ts t eer i ng ' ' (TD S) , works w e l l f o r a si ng le beam, butF i g . 2 shows th at op posi t ely charged beams f e e l equal

a nd o p p o s i t e f o r c e s i n t r a v e r s i n g a t i l t e d d o u b l e t; o n e

beam would b e s t e e r e d on to t h e ma ch ine a x i s wh i l e t h eo s c i l l a t i o n a m p l i t u d e of the o ther beam is i ncreased .

Th i s d i f f i c u l t y can be overcome i n t he fo l l owi ng

T h i s

manner.

i t y and consider a doublet which is FD f o r a p o s i t i v e l y

charged beam.

are shown i n F i g . 3. displacement of th e doublet f rom A t o B r e q u i r e s t h a t

equal and oppos i t e fo rce s be app l i ed t o t he quadrupo l es .

Wecan accompl ish th i s by applying equal and opposi te

Cons i der on l y t h e ho r i zon t a l p l ane f o r s i mp li c-

The for ce diagrams for th is ar rangement

It is e v id e nt t h a t a n e l e c t r i c a l

*Work performed under th e aus pic es of th e U. S. Atomic

Energy Commission.

New Mexico

s t e e r i n g f i e l d s t o t h e e nd s of t he doub l e t . Th i s is

c a l l e d "d i sp l aced doub l e t s t ee r i ng" (DDS) .

Now cons ider t h e f o r c e di agr am s (Fig. 4 ) f o r anega t ivel y charged beam. The f i r s t quadrupole i s now

D and th e second quadrup ole is F and t h e s t e er i n g f i e l d s

r e v e r s e t h e i r s i g n s . So w e s e e t h a t DDS e l e c t r i c a l l yco r r ec t s fo r mi sa l ignment s o f t he doub le t f o r nega t i ve l ya s w e l l as p o s i t i v e l y ch ar ge d bea ms. Hence t h e mostversa t i le s t e e r i n g c o i l c o n f i g u r a t i o n i s one which al-l ows i ndependen t d i po l e f i e l d s t o be superposed i n bo tb

p l ane s on each quadrupole. I f economics, or o ther con-s i d e r a t i o n s d i c t a t e less v e r s a t i l e c o n fi g u ra t i on s of

s t e e r i n g m a gn et s, o r t h e a s s o c i a t e d p o s i t i o n s e n s in g e-qui pment , ana l y t i c t echn i ques are e f f e c t i v e i n d e t e r -min ing t he bes t s t ee r i ng s t r a t egy . *

P r a c t i c e

A t LAMPF t h e r e a r e 134 quadrupo les i n t he Al varezl i n ac and 103 d o u b l e t s i n t h e s id e- c ou pl ed s t r u c t u r e .Since i t would r eq ui re two power sup pli es per quadru-

po l e f o r e l e c t r i c a l al ig nment, i t is r e a s on a b le t o i n-ves t i g a t e s t ee r i ng con f i gu ra t i ons wh ich min imize t r ans -

v e r s e o s c i l l a t i o n s f o r a small number of s te er in g posi-t i o n s . H enc e, t h e f i r s t g o a l d u r i n g c o n s t r u c t i o n is t oa t t a i n t h e b e s t d o u bl e t a l ig nm e nt p o s s i b l e .

l e t s i n t h e s id e- coup l ed s t ruc t u re are a l i g n e d t o 2.007 and 2 .4mr. Nu mer ic al s i mul a t ions show t h a t a

. O O

displacement has approximately the same e f f e c t as

a 0.25mr doub l e t ax i s tilt. The a l i gnmen ts i n t he s ide -coupled l i na c a re comparable to t hes e, and hence i t i s

necessary to have a combination of TDS and DDS.

The doub-

I n t he s i de- coupled s t ructure, one doublet per mod -

ul e i s wired to provide a combination of TDS and DDS.However , each doublet has s tee r in g i n only one plane.

Based on numeri ca l s i mu l a t i ons , whi ch a l so c ons i der t he

number and loc at i on of pos i t i on moni tors , i t has beenf ou nd t h a t s t e e r i n g i n one p l ane i n two s uc e ss i v e mod -u l es and t hen i n t he o t her p l ane fo r t h e nex t t wo mod -

ul es ( i . e . , t h e p a t t e r n is HHVV and then repeats) i s anef f ec t i v e con f i gu r a t i o n .

T he e f f e c t o f t h e e a r t h ' s m a g ne ti c f i e l d h a s a l s o be en cons ide re d. It is shown in ref ere nce 1 t h a t t h ee a r t h ' s f i e l d d i s p l a c e s t h e e q u il i br i um o r b i t .

men t a l r esu l t s i mp l y t ha t t h i s d i sp lacemen t is 0.15 cmi n t h e s i de-c o up le d s t r u c t u r e .is d i sp l aced equal l y and oppos i t e l y fo r oppos i t e l y

charged beams , hence t o mi n imize t r ans ver s e o sc i l l a t i o ns ,

it is n e ce s sa r y t o s t e e r so that the beams are pos i t ion-

ed at t h e i r r e s p e c t i v e eq u il ib r iu m o r b i t s r a t h e r t ha n a tt h e d e si g n c e n t e r of t h e l i n a c .

Experi-

The equ i l i b r i um o rb i t

I n t h e Al va re z l i n a c , t h e r e i s not enough posi t ioni n f or m a t io n t o e s t a b l i s h t h a t a n o ff a s i s equ i l i b r i um

orb i t ex i s t s . However , i t i s e xp ec te d t h a t t h e e f f e c to f t he ear t h ' s f i e l d would be small because of t h e s t e e li n t h e t ank wal l wh ich i s n o t p r e se n t i n t h e s i de-

coupled l inac.

If one wishes t o d o e f f e c t i v e s t e e r i n g , t h e p o s i t i o ni n fo rmat i on shou l d no t be separa t ed f rom t he s t eer i ng

529


20/226

magnet by too many in terve n ing quadrupoles . I f th ere

are many misaligned quadrupoles befo re t he pos it ion in-f o r ma t i o n , t h en t h e p o s i t i o n i n f o rm a t i o n is not very

u s e f u l . The b es t co n f i g u ra t i o n seems t o b e t o h a ve p o s i t i o n in forma t ion i n two s u cces s i v e cells a f t e r t h es te er ing magnet.

Acknowledgement--

I would l i k e to thank Don Swenson f or sug ges tin gx

f x

t h e u s e of t h e f o r ce d i agr ams .

\

\

References

1. D. A. Swenson and K. R. Cr an d a l l , Los AlamosSc ie n t i f i c Laborato ry , Pr i vat e Communicat ion,Ju ly , 1968 .

2. D. A. Swenson, Los Alamos S ci en ti f i c Laboratory,

Pr iv a te Communication, August 1968.

F q u a d , H + D quad, H-

FIG. I

f

X

f x

f X

f x

F quad, H t D quad ,H+ D quad,H- F quad,H-

FIG. 2

f x f x

f x f x

F q u o d ,

H+ D quad, H+ D quad, H- F quad, H-

FIG. 3 FIG. 4

53 0


21/226


22/226

V = - 2 MV v =o v = O V = + 2 M V

V = - 2 MVQ

+2MV

I /

--IONSOURCE

PIERCE ELECTRODE PIERCE ELECTRODE

ELECTR0NBEAM

SYSTEM

D-T ION BEAM

F i g . 1. Line ar co l l id in g beam sys tem.

where I i s to ta l curre nt (2000 amperes) ,pc

i s i o n v e l o c i t y ( a bo u t 1 . 5

eo i s t h e d i e l e c t r i c c o n st a n t

r o i s the radius of the beam.(1.1 x 10-10 F/m)

Approximate ly th i s force i s

2Er - vB = 1200 I r / r v o l t s /m

I f a r e l a t i v i s t i c e l e c t r o n beam

9

x 10')of f r e e s p a c e

( 5 )

of I, amperes i s

now introduced, co l l in ea r with the ion beam but t ra vel -

l i n g i n t h e o p po s it e d i r e c t i o n , i t w i l l c o n t r i b u t e afocus ing force of

where Pec i s t he e l ec t ron ve loc i ty (as sumed approx-i m a te l y e q u a l t o c ) . The i n t e r n a l f o r c e s i n a r e l a -t i v i s t i c beam approximate ly cance l each o the r and the

electron beam w i l l exper i ence a focus ing force due tothe ion beam of approximately the s t r eng th given by

Eq. ( 5 ) . I f the inward force s on ions and e lec t ro nsa r e set e q u a l, t h e e l e c t r o n c u r r e n t r e q u i r ed p r ov e s t o

be ab ou t 80 ,0 00 am pe re s.

Under the for ces j us t di scus sed, both beams w i l lcol l apse to a smal l d i amete r de te rmined by th e or ig -

i n a l v a l u e s o f t h e i r e m i t t a n c e s. I f w e assume

(opt imis t i ca l ly) an emi t t ance of 1 0 0 ~m.mrad f o r t h eion beam, the f in a l beam radius proves to be 1 . 4 nun.

The charge den si ty in each io n beam i s about

11 coulombs/m3 and the fusion power i s now 17 MW/m3.But t h e beam has become so s m al l t h a t t h e a c t u a l power generat ed per meter of beam i s b a r e l y over100 wat t s .

I n F i g . 1 we present a conf igu rat i on of e lect ro desfor t he l in ea r co l l id ing beam system. Thi s a r range-

ment makes poss ibl e th e gene rat i on of 2-MeV beams ofions and ele ctr ons and t he dece ler at i on of both beamsfo r recovery of the energy s tor ed . One can sa f e ly

conc lude f rom the pa ramete rs jus t presented tha t sucha system w i l l n e v e r b e b u i l t .

3 . Proposed Configurat ion

What evidently is requi red t o make a co l l id i ng beam syst em v i a b l e i s a method fo r s t or in g the ion beams u n t i l they i n t e r a c t . I€ t h i s c an b e do ne t h e

input io n cur ren ts become qu i t e reasonab le . For an

output of 1 W o f f u s i o n power a l l t h a t i s r e q u i r e d i san input of 60 each of deuterons and t r i tons.

We note fu r th e r tha t a pr ime requirement of thesys tem i s t h a t i t i n c l ud e s t r o n g r e s t o r i n g f o r c e s

which w i l l prevent coulomb s c a t t e r e d io n s fr om leaving

the sys tem before t h e y h av e ti me t o t a k e p a r t i n af u s i o n r e a c t i o n .

The sys tem t o be proposed i nclu des coin cide ntdeu tero n and t r i t o n beams of t he same momentum cir cu la -

t i n g i n a p pr o xi m at e ly c i r c u l a r p a t h s i n a r a t h e r h i g hmagnetic f i e l d, focused by a cyl in dr ic al beam of elec-

t r o n s t r a v e l l i n g a l o n g th e l i n e s o f f o r c e of t h e f i e l d .F igure 2 i s a c r o s s- sec t i on ske tch of the geometry .

To s a t i s f y t h e r e l a t i v e v e l o c it y c r i t e r i o n and t o

have the same momentum the deuteron energy must be

845 keV; t h e t r i t o n e n e rg y w i l l be 564 keV.t e r o n v e l o c i t y w i l l be 9 x lo6 m/sec; t h e t r i t o n v e lo c

i t y w i l l b e 6 x l o6 m/sec. I n a f i e l d of 6 t e s l a , t h erad ius of cur vatu re of these beams w i l l be 3.0 cm.

The deu-

B

i

iiiii

CATHODEJ

( V = -v,

1

ANODE (V.0)

PAR AX I ALELECTRON BEAM

AND AZ IMUTHA LLYCIR CULATl NGION BEAMS

E

PLATE ATCATHODEPOTENTIAL

F i g . 2 . C r os s s e c t i o n t hr ou gh c y l i n d r i c a l c o l l i d i n g beam sy st em .

4 . Dynamics o f th e E le ct ro n Beam

We c o n si d e r f i r s t t h e be h a v io r o f t h e e l e c t r o n

5 3 2


23/226


24/226

have assumed (Bz = 6 T, p = 11.7 coulombs/m3), t h i s

d i m e ns i o n le s s q u a n t i t y h a s t h e v a l u e 0 . 2 1 and t he valueof 6 giv en by (19) is 0 .26 6 1 - .1 .26 62 . I f 61 = 62,t h e o t h e r v a l u e o f ?jfor which 6 = 0 i s approximately

- 6 1 -

F or t h os e f a m i l i a r w i t h t h e no':ation of plasma

physics , t h e q u a n t i t y ( 4 n p / € ) m/eBz) can be recognized

as an ana log of the p l a sma @ funct ion which is a meas-u r e of t h e r a t i o of plasma p r es su r e t o ma gn et ic pres-sur e. In a plasma @ must be kept below unity.

5. Dynamics of th e Deuteron and Tr it o n Beams

We a ss um e, i n i t i a l l y , t h a t a small number ofd e u t er o n s and t r i t o n s a r e i n j e c t e d i n t o t h e sp ace-

c h a r ge f i e l d c a l c u l a t e d in t h e p r e c e d i n g s e c t i o n f o r

t h e e l e c t r o n s h e e t . T he se i o n s a r e t o move i n a f l a ts p i r a l w i th n e g l i g i b l e ve l o c i t y i n t h e z - d i r e c t i o n .

The method of in j ec t io n in to th i s o rb i t w i l l bed e s c r i b e d i n t h e n e xt s e c t i o n .

Motion of the ions w i l l be go ve rne d by

2m.v

where m i f s t he ion mass,v i is t h e i o n v e l o c i t y , given by mivi = - Bzero ,

Er

is given by (13) above.

In the coordina tes of the preceding sec t i on (20) becomes

6 . I o n I n j e c t i o n

Deuterons and t r i t o n s a r e t o b e i n j e c t e d i n s u ch af a s h i o n t h a t t h e y w i l l c o nt i nu e t o c i r c u l a t e i n t h e

magnet ic f i e l d and w i l l be unable t o es ca pe.i n j e c t i o n t h e y w i l l be given a s l i t t l e a x i a l momentum

as p o s s i b l e . Esca pe of ions a t th e ends of t h e d e v i c e sw i l l be prevente d by a l o c a l i n c r e a s e i n m a gn e ti c f i e l d

The l o c a l i n c r e a s e i n a x i a l f i e l d w i l l be accompanied by in t r o d u c t io n of a r a d i a l f i e l d component which w i l l

se rve to rever se the pa raxi a l ve loc i ty of the ion beam.Fiel d bumps of t hi s type w i l l be inclu ded a t bo th en dso f t h e d e v ic e a s i n d i c a t e d i n F i g . 2 .

During

/

INFLECTORyj\

.,._.... ......

CLOUD

whence

whose solut ion i s

D l AN D T

BEAM INJECTEDD AN D T Z

I O N B E A M

6

and cp a r e d et e rm i ne d b y i n i t i a l

c o n d i t i o n s .

We n o t e t h a t , s i n ce p r e p r e s e n t s t h e d e n s i t y o f a ne le c t r on space charge , the two terms i n w2 bo th ha ve

t h e same s ign . For deute rons

9w = ( 8 . 3 3 x + 6.33 x = 7.96 x 10 .

The wavelength of this " b e ta t r o n o s c i l l a t i o n" i s 0.71 mm( f o r d e u t er o n s) .

b e t a t r o n wa ve le ng th is 0.87 mo s c i l l a t i o n h a s i t s frequency determined almost com-

p l e t e l y by t h e d e n s i t y of th e e l e c t r o n i c sp ace charge .

F o r t r i t o n s , w = 6.50 x l o 9 and the

Thi s very sh ort wave

The ve ry s t rong res tor ing f orce provided by thee l e c t r o n c l o u d s ho ul d be e f f e c t i v e i n r e s t o r i n g t o

th e i r or b i t s , i ons th a t have undergone coulomb sca t -

t e r i n g e i t h e r by o t he r i o n s o r by e l e c t r o n s . T h is

t o p i c is not ana lyzed i n th i s r epor t bu t mus t be givena t t e n t i o n i n f u t u r e s t u d i e s of t h i s d e v i c e .

F ig . 3 . I n j e c t i o n f o r c y l i n d r i c a l c o l l i d i n g beamsystem.

Severa l me thods of in j ec t ion w i l l o cc ur t o t h e

r e a d e r . One p os s ib l e method i s i l l u s t r at e d i n F ig . 3 . This method u t i l i ze s molecular ions which pass through

a n i n f l e c t o r t o b e d ef l e c t e d o nt o an o r b i t t h a t i n t e r -s e c t s t h e e l e c t r o n c lo u d .

10%of th e molecular beam should be s t r i p p e d b y e l e c t r o nc o l l i s i o n s t o become at om ic i o n s wh ich then w i l l swi tch

to o rb i t s through the el ec tr on cloud (we assume a

s t r i p p i n g c r o s s s e c t i o n o f t h e o r d e r o f 10-16 cm2).remainder of th e molecular ions w i l l c o n t i n u e o n c i r c u l a ro r b i t s and r e t u r n t o t h e i n f l e c t o r . To p r ev e nt t h e i r

l o s s by a seco nd d e f l e c t i o n , they w i l l make t h e i r f i r s ten t r y in to t he inf l ec to r wi th a small component of

p a r a x ia l v e l o c i t y . The i n f l e c t o r is to have a f i n i t ea x i a l e x t e n t and t h e p a r a x i a l v e l o c i t y o f t h e ions w i l l

be su ch as t o a l low the beam t o m i s s t h e i n f l e c t o r on

t he s econd and l a t e r revolu t ions . Thus the molecula r

Something of the orde r of

The

beam w i l l r e- e n t e r t h e e l e c t r o n c l o u d s e v e r a l times

534


25/226

un t i l v i r t u a l l y a l l of t he beam has been r educed toatomic ions. This scheme has the v i r tu e of al lowingcon t i nuous i n j ec t i on . Pu l sed i n j e c t i on p rocedures may,

however, prove t o be simpler and le s s demanding ofex t r a magnet ic f i e l d vo lume.

Another pos sib le i nj ec ti on method would involvei n j ec t i o n o f n eu t r a l a toms p roduced by a c c e l e r a t i o na n d s t r i p p i n g of nega t i ve i ons . These t echn i ques a r ew e l l e s t a b li s h e d f o r u s e i n i n j e c t i o n i n t o t h e AGS.The ne ut ra l atom beam would be i n je cte d tangent t o the

e l e c t ro n c l oud where a f r a c t i on o f t he o rde r o f 10%would be ionized and proceed on t h e d e s i r e d c i r c u l a ro r b i t s .

The p rocedure fo r i n i t i a l l y combin ing t he deu t e ron

and t r i t o n beams i n t o a s i ng l e beam i nvo l ves e l e c t ro -s t a t i c de f le ct ion . The two beams, having the samemomenta but d if fe re nt en er gie s can be combined by de-f l e c t i o n i n an e l e c t r o s t a t i c f i e l d .

7 . Procedure with High Density Lon Beams

The preceding sec t ion s de al t wi th the mot ion ofion beams of low in t en si ty i n a dense shee t of e lec -t r o n s . It h a s b e en shown t h a t , t o t h e f i r s t o r d e r ,the motion of both ele ct ron s and ions i s s t a b l e .There a r e l a rg e r es t o r i ng fo rce s on t he i ons wh ich canserve t o coun t e rac t t h e undes i r ab l e e f f e c t s o f cou lomb

s c a t t e r i n g .

I f , now, t h e e l e c t r o n d e n s i t y i s doubled and the

i o n d e n s i t y i s r a i s e d t o t h e l e v e l of t h e o r i g i n a le l e c t r o n d e n s i t y , t h e n e t d e n s i t y and t h e e l e c t r i cf i e l d p a t t e rn w i l l be unchanged. Onl y the d i s t r i b u t i o n

of B, w i l l be a f f e c t e d by t h e c i r c u l a t i n g i on c u r r e n t .For t he den s i t i es quo t ed , B w i l l drop by about 0 . 1 Tt h rough t he t h i ckne ss . Th i s d rop i s t o o s m a l l t oa f f e c t p e r c e p t i b l y t h e e l e c t r o n o r i o n m ot io ns .

The procedure t o incr ease de nsi ty would be t ora is e the in jec ted ion cur ren ts . The pote nt i al maximum

i n t he e l e c t ro n shee t woul d t hen d rop and t he e l ec t roncu r ren t supp l y wou ld au t omat i ca l l y add e l e c t ro ns t orestore the maximum value of the potent ial .

When the io n dens ity ha s reached 11 .7 coulombs/m3,t h e f u s i o n r e a c t i o n s w i l l yield 3600 wat t s of f u s i o n

power per me te r le ngt h of the sys te m. The deu t e ronand t r i t o n supp l i es a r e r equ i r ed t o p rovi de on l y abou t200 @ e a ch t o m a i n ta i n t h i s y i e l d .

It would appear that the procedure of pushing ioncu r ren t and e l ec t ron c u r r en t up , mai n t a i n i ng a cons t an td i f f e r e nce be t ween t h e i r charge dens i t i es , can be con-t i n u e d i n d e f i n i t e l y t o y i e l d h i g h e r and h i g h er l e v e l s

of fusi on power . No doubt , however , in st a b i l i t i e s w i l l pu t a s t op t o t h i s . The po i n t a t which t h is happens

w i l l be d i f f i c u l t t o p re d i c t t h e o r e t i c a l l y and mi gh tmore ea si ly be determined exper im ental ly .

DISC USSIO N

V. Ke lvin Neil (LL L):

elect rostat ical ly ra the r than magnet ical ly?

Blewett: Yes.

V. Kelvin Neil:you're trying to get produces the helium?

Blewett : Yes.

V. Kelvin Neil:held? I 'm trying to get to one of the problems in TOKOMACwhere the hel ium is contained and, in effect , quenches thereaction.

Blewett (restat i ng thequestionk:

Your atoms are held in the device

And each one of these reactions which

And the helium is also then elect rostat ical ly

The containment syste m

is essent ial ly a n elect rostat ic containment system and thatone of the products of the reaction would be a helium ionand will the helium ions do the same po is on in g of the reac-tion as they do in TOKOMAC r e a c t o r s ? I don't think I cangive a very good answ er to that , except to say that thehelium ions have about 4 MeV of energy which should beenough to kick th em out of this region.

LeonKatz

(Universi ty of Saskatchewan): pose d as a sou rce of energy or as a source of neut rons for b r e e de r s ?

Is this being pro-

Blewett:should say that it is be ing pr op os ed only as a n experiment.

-4rie

Van Steenbergen (BNL):acce l e r a t o r s i s no t an establ ished fa ct .

These are sort of in terchangeable, aren ' t they? I

Charge exchange injection of

5 3 5


26/226

WILL NEGATIVE HYDROGEN I ON SOURCES SOON REPLACE PROTON SOURCESI N HIGH E NER GY ACCELERATOR S?"

Th. S lu yt er s and K. P r e l e cBrookhaven National Laboratory

Upton, New York

Although charge exchange in j ec t i on in to c i r cu l ar

ac ce le ra to r s and s to rage r in gs has been p roposed qu i t esome t i m e a go , t h e a p p l i c a t i o n o f t h i s a t t r a c t i v e m et h-od has not been widespread because of low in te ns i t ie so f nega t ive hydrogen beams ava i la b le un t i l r ece n t ly .For s y n chr o t r on s and s t o r ag e r i n g s m u l t i t u r n i n j ec t i o no f p r o t on s v i a s t r i p p i n g of n eg a t i v e i o n s o f f e r s a

b e t t e r and s impler a l t e r n a t i v e t o t h e p r e s en t i n j e c -t i on schemes by incre as ing t he phase space dens i ty o fth e coa s t i ng beam. For cyclo t rons in je c t io n o f pro-t o n s v i a s t r i p p i n g o f r e l a t i v e l y h i gh e n er gy n e u t r a l part icles ( o b t ai n ed by p a r t i a l s t r i p p i n g o f n eg a t i v eions ) may a l le v i a t e th e space charge p rob lem dur ing th eea r l y par t o f acc e le ra t ion . However, beam in te ns i t i e s

o f nega t ive hydrogen ions ob ta ined by d i r ec t ex t r ac t io nfrom stand ard sources such as duoplasmatrons and Penn-ing sources were seldom h igher than seve ra l mi l l i amperes , wh ic h was n o t s u f f i c i e n t fo r most o f p resen t c i r -cu la r acc e le ra t o r s . I nd i r ec t method o f p roducing nega-t i ve io n beams vi a charge exchange of protons , al thougha t t h a t t i m e pr om isi ng wi th respect t o t h e i n t e n s i t y ,

had a disadv ant age of yie ld in g beams of a too low qua-l i t y and r equ i r ing a too complex mechan ica l s t ruc tu re .

D ur in g t h e l a s t y ea r o r so severa l paper s and re- p o r t s appeared d e s c r i b i ng two new appro aches t o t h e p roduct ion of nega t ive hy drog en beams e x t r a c t e d d i r e c t -l y f rom a pl as ma . One o f them was t h e ho llo w di schargeduoplasmatron,

4 6 developed f rom a s ta nda rd so urce by p l ac i ng a rod a long th e main ax i s and r each ing in t o the

anode d i scharge r eg ion .was obtai ned with a normalized emit tanc e less than 0.1cm-mad. An accompanying ele ct ro n cur re nt of 0.5 A, ar e l a t i v e l y l o w i o n cu r r en t d en

Firing Angle Document

Documents

Transcript of Firing Angle Document